We propose a systematic and quantitative study of the structural discontinuities and instabilities at phase boundaries of lipid bilayers and their effects on protein activity in reconstituted membranes. We will use a combination of x-ray diffraction, 31P NMR, calorimetry, ultra-rapid freezing and freeze etching, low dose diffraction contrast imaging of hydrated specimen and analytical electron microscopy to obtain information on the state of phase separation in binary and ternary mixtures of natural phosphatidylcholine, phosphatidylethanolamine, phosphatidylserine, phosphatic acid, cardiolipin and cholesterol. The conditions for cation-induced phase separation will be determined, and non-bilayer structures as a particular form of instability will be examined. Computer-aided statistical analysis and pattern recognition will be used to quantitate the pictorial information. Relations between domain geometry and morphological changes (fusion) will be studied. Kinetics of these structural changes will be measured by the stop-flow method in conjunction with rapid freezing. Incorporation (into controlled lipid mixtures) of erythrocyte membrane proteins and ATPase will be evaluated. Binding of cytochalasin-B to the glucose carrier proteins of erythrocytes and the ATP exchange rate in the reconstituted membranes will be assayed under conditions at which lipid structural changes are expected to occur. The reported dependence of the activities of these proteins on lipid composition will be examined in the light of structural discontinuities and instabilities. The structure-activity correlation will give us more insight into this new and less-known aspect of membrane behavior.